A busy week of Big Tech earnings left us feeling more assured about our investments in Nvidia , Broadcom and Eaton as spending on artificial intelligence chips and data centers remain clear-cut priorities. The quarterly reports from Club holdings Alphabet , Microsoft , Meta Platforms and Amazon this week varied in overall quality, but a common thread woven across them all was a continued commitment to big investments in AI. More crucially, all four companies indicated their investments — reflected in their capital expenditures, or capex, outlooks — were picking up steam. Capex is money spent to buy or upgrade physical assets, including the construction of data center buildings and the computer hardware inside them. Nvidia, Broadcom and Eaton are poised to benefit from this ongoing spending: Nvidia as the dominant maker of AI chips; Broadcom as a partner with Alphabet on the Google parent’s proprietary AI chip and a provider of data-center networking technology more broadly; and Eaton as a supplier of electrical components and power systems used to run data centers. Among the three, Nvidia is the biggest winner, given how much of its overall revenue is tied to AI initiatives, followed by the more diversified efforts from Broadcom and Eaton. Nevertheless, Broadcom and Eaton’s exposure to the fast-growing AI market is part of our investment cases for both firms. For its part, Eaton’s earnings report Thursday morning fortified our thesis, offering an optimistic 2024 outlook that includes strength in its data center business. Nvidia is scheduled to report quarterly results on Feb. 21, while Broadcom is expected to follow shortly thereafter. Big Tech spending AMZN 1Y mountain Amazon’s stock performance over the past 12 months. Amazon’s capital expenditures are expected to increase year over year in 2024, CFO Brian Olsavsky said Thursday. The primary drivers of the uptick will be infrastructure investments to fuel growth in cloud-computing unit Amazon Web Services (AWS), the CFO said, including initiatives on generative AI and large language models. Amazon’s incredible quarter released Thursday evening was powering its stock up 8% on Friday Amazon did not provide specific capex guidance for 2024, but analysts are modeling for around $60 billion, according to estimates compiled by FactSet. Amazon’s capex actually fell about 17% year over year in 2023 to $48.4 billion, as the company pared spending on its e-commerce logistics network. Some of Amazon’s spending is likely to go toward its custom AI chips, known as Trainium and Inferentia. But the Seattle-based tech giant also buys Nvidia chips. On Thursday’s post-earnings call, CEO Andy Jassy said AWS offers the “most expansive collection of compute instances with Nvidia chips.” In November, the two companies announced an expanded partnership that brings Nvidia’s supercomputer service DGX Cloud to AWS. META 1Y mountain Meta Platforms’ stock performance over the past 12 months. Alongside its knockout earnings report late Thursday, which drove shares more than 20% higher Friday, Meta Platforms increased the upper end of its full-year capital expenditures forecast to $37 billion from $35 billion. The Instagram and Facebook parent left the low end of its guidance unchanged at $30 billion. In 2023, Meta’s capex totaled $27.3 billion, down from $31.4 billion a year earlier. Driving the expected growth in 2024 is spending on servers — both AI-specific servers containing AI chips like Nvidia’s and those meant for more general computing — and construction of data centers designed to better handle the needs of AI workloads, CFO Susan Li said on the earnings call. “While we are not providing guidance for years beyond 2024, we expect our ambitious long-term AI research and product development efforts will require growing infrastructure investments beyond this year,” Li said. CEO Mark Zuckerberg reiterated that Meta is on track to own 350,000 of Nvidia’s top-of-the-line AI chip, known as the H100, by the end of the year. Factoring in other processors, Meta’s AI-focused computing infrastructure will be equivalent to almost 600,000 H100s, Zuckerberg said. Meta has said it’s also ordering the new AI chip from Advanced Micro Devices , the MI300X, which launched late last year as an alternative to Nvidia. The social media behemoth has plans to use custom chips for certain AI tasks, too. MSFT 1Y mountain Microsoft’s stock performance over the past 12 months. Late Tuesday , Microsoft — one of the biggest Nvidia chip buyers lately — said it expected capital expenditures in its current quarter to “increase materially” compared with the three months ended Dec. 31. In the prior period, which was Microsoft’s fiscal 2024 second quarter, the company spent $9.7 billion on property, plant and equipment. Wall Street sees that totaling $11.64 billion in the current quarter, according to estimates compiled by FactSet. “These data center investments support our cloud demand, inclusive of needs to scale our AI infrastructure,” CFO Amy Hood said on the call. Microsoft’s past spending has already translated into financial gains — its cloud-computing unit Azure grew 30% on an annual basis in its fiscal second quarter, with AI services notably contributing six points of the growth. In fact, Azure’s growth rate exceeded that of AWS and Google Cloud for the second consecutive quarter. It’s further evidence of Microsoft’s emergent leadership position in AI, rooted in its close partnership with OpenAI, the startup whose AI chatbot ChatGPT went viral in November 2022 and sparked the ongoing generative AI investment wave . Microsoft also has placed orders for AMD’s MI300X. GOOGL 1Y mountain Google parent Alphabet’s stock performance over the past 12 months. Alphabet expects its capital expenditures in 2024 to be “notably larger” than in 2023, which totaled $32.5 billion, finance chief Ruth Porat said on Tuesday evening’s earning s call. Alphabet didn’t offer specific figures, but analysts currently project full-year capex of about $41 billion, representing 26% year-over-year growth, according to FactSet. In the company’s fourth quarter, capital expenditures rose 37% sequentially to $11 billion, driven largely by spending on servers followed by data centers, Porat said. Alphabet’s spending on AI chips has typically been split between its custom chips designed in partnership with Broadcom, known as Tensor Processing Units (TPUs), and Nvidia’s offerings. In this way, both Broadcom and Nvidia are poised to benefit from Alphabet’s server spending. Back in August, Alphabet announced its fifth-generation TPU, which the company has traditionally used for its internal AI work. At that time, Alphabet also detailed an expanded partnership with Nvidia that brought its supercomputing DGX service to Google Cloud. For all of Broadcom’s fiscal 2023 — numbers released with Q4 earnings in December —the company generated roughly $2.8 billion in “AI accelerator” revenue, which is essentially codename for TPUs. Including networking products that stitch together parts of the data center, about 15% of Broadcom’s semiconductor revenue was tied to generative AI spending in fiscal 2023. The company expects that to rise to more than 25% in its current fiscal 2024. (Jim Cramer’s Charitable Trust is long NVDA, AVGO, ETN, MSFT, GOOGL, META and AMZN. See here for a full list of the stocks.) As a subscriber to the CNBC Investing Club with Jim Cramer, you will receive a trade alert before Jim makes a trade. Jim waits 45 minutes after sending a trade alert before buying or selling a stock in his charitable trust’s portfolio. If Jim has talked about a stock on CNBC TV, he waits 72 hours after issuing the trade alert before executing the trade. THE ABOVE INVESTING CLUB INFORMATION IS SUBJECT TO OUR TERMS AND CONDITIONS AND PRIVACY POLICY , TOGETHER WITH OUR DISCLAIMER . NO FIDUCIARY OBLIGATION OR DUTY EXISTS, OR IS CREATED, BY VIRTUE OF YOUR RECEIPT OF ANY INFORMATION PROVIDED IN CONNECTION WITH THE INVESTING CLUB. NO SPECIFIC OUTCOME OR PROFIT IS GUARANTEED.
A busy week of Big Tech earnings left us feeling more assured about our investments in Nvidia, Broadcom and Eaton as spending on artificial intelligence chips and data centers remain clear-cut priorities.
Read the full article here
